capitalized time
@capitalizedtime
Followers
849
Following
11K
Media
404
Statuses
2K
How to maintain privacy in today’s world . Redact your photos by using pixeldrop by @pugson!!. 1. 1 click micro interaction for auto-[REDACTING] faces + text.2. Easily fine tune the REDACTION by adjusting the box handles .3. Tweak the REDACTION style to blackout, glitch, or large.
Pixeldrop by @pugson is [REDACTED REDACTED REDACTED REDACTED] for [REDACTED] sensitive personal info in your [REDACTED REDACTED] photos. Hope you're picking [REDACTED] what I'm putting [REDACTED].
2
0
1
I also have full detailed information architecture flows of all 18 LLM apps~. Let me know if you'd like me to open souce these. Are there other LLM product experiences you'd like a deep dive on?
0
0
0
6/ The full 49 page deck is available here with comparisons across 18 LLMs and LLM experiences. @OpenAI @ChatGPTapp @Alibaba_Qwen @grok @ManusAI_HQ @deepseek_ai @GeminiApp @AIatMeta @MistralAI @Copilot @AnthropicAI @perplexity_ai @Kimi_Moonshot @youdotcom @huggingface @pi.
1
0
0
5/ Real time voice - voice is 3x faster than typing (from @WisprFlow). Building a good voice product comes down to the vision of the future that you envision. Jarvis from Ironman, Her by Spike Jonze, C-3PO by Star Wars are just a few narratives crafting the future of voice. It
1
0
0
4/ The future microinteractions are seamless ChatMessage with Actions. What do we do with the chatResponse generated from the AI assistant? Copy paste the code, feed into another app, acknowledge it, reference the store of truth? With a click or tap, this can be disseminated
1
0
0
3/ Tool use. As Jobs said, "a computer is a bicycle for the mind". The future of the LLM interactions is 100% going to be personalization of all the incredible tools callable by LLMs. @TONGYI_SpeechAI blows the rest of the LLMs out of the water with 11 tools. How to steal this
1
0
0
2/ Default prompts - a single tap allowing users to go through a happy path preconfigured to the differentiated value prop of your product. We live in post-scarcity society, there are easily 100+ LLM app interfaces, why should a user use yours? Make it simple to see the core
1
0
0
1/ For multi-modality, users can input anything from PDF documents like Circle's S-1, to a photo from their library, or an irl photo with their camera. How to steal this idea:.Start with determining the inputs that your user will input. If it's text, audio, photos, files, then
1
0
0
Steal the best human computer interactions for LLMs from Gemini, ChatGPT, Tong Yi, Perplexity, Manus, Meta AI, and more. What are the best experiences across multi-modality, default prompts, tool use, chatMessage with actions, and real time voice?. Here's the breakdown:
1
0
1
If you don’t have a visual artifact, you don’t have a movement either. Cue all the symbols from “Iron Fists Branding - The 20th Century Totalitarian State” to “Logo Modernism”.
If you don’t have a slogan, you don’t have a movement.
0
0
1
what a great url slug. no ai, no io. just 2 endearing pals.
0
0
0
Noam Shazeer's accomplishments and timeline. 1. co-author of "Attention is All You Need".2. developed a chatbot called Meena.3. left google when they refused to release it.4. launched 5. bought by Google for $2.7 billion .6. working for google again
Whoever’s doing comms for Gemini needs to apologize to Noam Shazeer and then resign. DeepMind doesn’t get nearly as much credit and attention as they deserve, and that’s because comms is vastly underperforming capabilities. You’ve got Demis, Noam, Jeff Dean, and Sergey Brin all.
1
0
3
who will be the one leading the charge for agents, agents, agents, agents in 2025?. is the new .NET conference Openai devday, Code with Claude, Llama con, WWDC or something else?
1
0
2
. @doriandargan, would a strategic alliance with Cash App or @AWGE be feasible for introducing Doji to the world?. In the wake of A$AP Rocky and Rihanna’s Met Gala appearance—coupled with their Cash App / Afterpay merchandise launch and announcement of their third child—the
0
0
1